Intense volatility and dispersion continued in what is normally a quieter month due to pre-fiscal year end earnings guidance which kicks off the day before the Golden Week holidays begin. Despite the geopolitical uncertainty which could create demand shock later in the year as well as long-term structural changes to energy supply chains, defense spending, and international trade, the markets appear to assume that, for now, there will be no systemic spillover and inflation will be contained, or at least manageable. As such, the markets have further focused on the one secular growth story that is, at least for now, unaffected by geopolitics – AI. Just like the Nasdaq (+15.3% in April) strongly outperformed the S&P 500 (+10.4%) and the Dow Jones (+7.1%), the Nikkei 225 (+16.1%) blew past the broader TOPIX (+6.6%). The Nikkei 225 isn’t a tech index but ever since the controversial “Big Bang” index rebalance of April 2000 when they replaced 30 old economy stocks with technology and telecom names in a single day, it has become more and more like one, especially in recent years. For example, looking at the index attribution last month, the top 7 names accounted for 75% of the index performance. These names include 2 SPE companies (Advantest and TEL), 3 tech hardware plays (TDK, Ibiden, and Kioxia which they just added in April, doubling upon entry after already rising 9x in the preceding 7 months), and Softbank (the leading AI investor in Japan). The only name that wasn’t tech was Fast Retailing, the maker of Uniqlo brand clothing and the largest index constituent, at least in April (it has since been overtaken by Advantest in May). If I were to extend the list by another 6 names, they’d cover 90% of the index return (and 4 of the 6 are also tech names while the other 2 are high-beta machinery companies).
 
And it is this theme that drove the Nikkei 225 to the 2nd largest monthly return since the Apr 2000 rebalance (the 1st largest move being last October). Of course, the broader market didn’t fare badly at +6.6%. There was some large cap bias but that’s probably due to the fact that the major tech firms tend to be large cap, though I’d note that Kioxia, the 3rd largest global NAND producer, is still in our mid cap benchmark, which I’m sure will change at the next rebalance seeing that Kioxia is now among the largest 50 companies in Japan and whose earnings are set to overtake that of Toyota’s this fiscal year.
 
As mentioned in the last few monthlies, we had reduced some of our tech hardware exposure in order to fund our increasing exposure to software names to take advantage of what we feel to be an overreaction to SaaS-pocalypse. While we markedly benefited from that rebalance in March, our strong outperformance in April was driven by our position in Harmonic Drive Systems (“HDS”). HDS is the largest global manufacturer of strain wave speed reducers. HDS commercialized the (now trademarked) Harmonic Drive over 50 years ago and has grown its global footprint as industrial robots were developed, mass-produced, and advanced. While I’ve known the company for decades, our interest with the stock began over a year ago after observing many machinery companies seeing their orders bottom (though not yet recovering) after the post COVID inventory build and its subsequent decline as the overstocked channel inventories were slowly worked down. We noted that firms that produced components with especially high market share, like HDS, were experiencing longer recover times from the excessive order backlogs. HDS had yet another problem in that they had expanded capacity due to a new potential demand driver coming from humanoid robots, causing an operating loss for 4 consecutive quarters and little recovery since. While we would not normally invest in such a highly cyclical stock, it was still trading at cyclical lows but, unlike its peers, had a structural, multi-decade growth story that perfectly fit one of our long-term theses and yet was being ignored by the market due to short-term, non-structural earnings weakness. Furthermore, after much time spent with senior management and their (half-empty) production facilities, we believe they can maintain their lead in cutting edge strain wave speed reducers. They also benefit from the fact that they are the only mass producer with a manufacturing footprint in the US while their leading competitor is Chinese. HDS supplies both countries as does its competitor, but HDS has a significant advantage should humanoid robots and key parts in its supply chain become classified as strategic goods subject to CFIUS/FDI regimes. HDS’s current exposure to humanoids is miniscule with the bulk of sales going to applications such as industrial robots (including cobots), semiconductor manufacturing, medical equipment, aerospace, and general machinery. However, using even the most conservative estimates of humanoid robot growth, sales to humanoids will likely overtake current total sales well within the next decade and the cyclical portion of their revenue will become noise. We are now the 3rd largest shareholder according to Bloomberg and is now our largest position in our concentrated portfolio (second largest in our diversified portfolio).
 
While I personally don’t share the market’s dismissal of the Iranian conflict and believe it will lead to both demand decline and margin pressure especially in Japan, I feel secure knowing that our investments target long-term themes unjarred by its effects. In fact, I believe it will accelerate many of our long term themes while potentially creating new ones. Still, I do hope for a peaceful resolution soon.​​

​

Masaki Gotoh​​​

​

===

 

​​“Any presentation using the word EBITDA … you just substitute the phrase ‘bull**** earnings.’” – the late Charlie Munger at the 2003 Berkshire Hathaway annual meeting.
 
I remember, over a decade ago when they first listed, Groupon chose to disclose a non-GAAP KPI called “Adjusted Consolidated Segment Operating Income” or ACSOI. This was operating income ex customer acquisition costs, stock-based compensation, and acquisition-related costs. This turned their annual net loss into a profit. I remember laughing out loud when I first read of it. It says that Groupon was profitable if they do not plan on growing any customers nor paying their employees (since I assume a large proportion of their labor costs were stock-based at the time). While they ultimately scrapped it and have since replaced it with “traditional” adjusted EBITDA where they still adjust for stock-based compensation (which I suspect is much lower now than pre-IPO), they make other non-recurring adjustments like restructuring costs and asset impairments, both of which are traditional. However, there have since been more egregious KPIs such as WeWork’s infamous “Community Adjusted EBITDA” where they excluded marketing, development costs, rent and general admin which, to me, sounds pretty close to revenue especially since their business was that of developing and marketing rented real estate. If there is an aspect of accounting that is “fun”, these colorful expressions are it.
 
Now I can’t say that I completely agree with Mr. Munger and EBITDA does serve some purposes but only if one understands what it represents which is different by industry and even by company. For example, the removal of Interest disregards how the business is financed in order to compare similar businesses with different capital structures. The Tax, similarly, disregards differing tax regimes. And the Depreciation/Amortization ignores past capex or acquisitions and tries to estimate how much the existing business generates cash. Of course, to be precise, one should also include changes to working capital unless the existing business is not expected to grow.
 
But if the business needs to spend to grow whether it be through capital expenditures or acquisitions, it is asinine to exclude the DA. Similarly, of course, taxes and interest payments matter to cash flows so the CFO should think about tax optimization and their capital structure, taking into account the risk and potential returns to any changes of either. Even in a relatively high corporate tax regime like Japan, there are many ways to reduce the tax burden, especially with the many subsidies and tax incentives that our government provides such as that for R&D, wage hikes, and regional project funding. The latter two are heavily used by one of our companies, Otsuka Shokai, allowing their effective tax rate to be consistently below 30% despite being a 100% domestic business (and yet, sell side analysts typically use the standard 30~35% in their forward forecasts).
 
So, one must be careful when dealing with EBITDA (or adjusted EBITDA) and ask, “what exactly is this measuring?” But I find that many people in the industry haphazardly use it. EV/EBITDA multiples are extremely common, particularly in private equity, which I find interesting given that business growth should be an important variable when thinking about the businesses they acquire. A skewed view may be that they want to hide the fact that interest burden has risen from the financial engineering they forced during the leverage buyout and that the growth, while it was private, was driven by acquisitions and other expansions with questionable recurring ROIs (in addition to the value destroying cost cuts to further improve EBITDA).
 
I similarly have problems with the standard IFRS methodology and is probably why firms find the need to use adjusted metrics (whether it be adjusted EBITDA or adjusted EBIT), particularly because many non-recurring expenses are incurred above the operating line. Also, IFRS adopters often see higher EBIT due to R&D accounting. And with IFRS 16, EBITDA has become even further inflated vs J-GAAP and US GAAP. Rent expense has become extremely difficult to analyze using Japanese financials even after going through the notes since lease disclosures are not standardized. I do think that the elimination of amortization of goodwill is a good idea though one needs to scrutinize balance sheet goodwill more carefully rather than to assume goodwill will eventually “go away” since it is amortized under J-GAAP. To make matters worse, Japan does not mandate a single metric although the trend is to move toward the least transparent IFRS. You can use J-GAAP, IFRS, Japan-modified IFRS (rare), or US GAAP. If I were a CFO, I would, of course, use the one that makes my EBIT (or EBITDA given the market’s infatuation with that metric) look highest, which would most likely be IFRS. Perhaps that is why only 10% of listed companies but 50% of companies by market cap use IFRS, i.e. the large cap companies (with savvier accounting teams and, arguably, the more stock price sensitive management teams) are converting to IFRS more quickly.
 
I believe J-GAAP is cleanest as it is very rules-based and differentiates between operating income (income from core businesses), recurring income (operating income + recurring financial income like interest, dividends, and FX), and pretax income (which includes non-recurring gains and losses such as asset disposals, impairments and restructuring costs, non-recurring securities gains/losses, litigation expenses, and disaster-related compensation or costs). But even with this regimented regime, “non-recurring” can be arbitrary. For example, retailers often write off post-season inventory under extraordinary losses. But most retailers have inventory write-offs every year and whose volatility often differentiates a strong or well-managed retailer vs one that is not. I generally add those losses back in the recurring income line and, for my forward estimates, calculate some percentage of inventory that will become a future loss (which could, potentially, be a point of engagement). And to do a proper DCF analysis, one cannot ignore working capital if the company expects to grow, a figure frequently neglected by both management and investors, which I find odd when investors push companies about use of future cash flows, not doing a proper analysis of cash flows themselves.
 
Now I wouldn’t mind if all investors understood the distinctions. But sadly, very few do, and that includes the sell-side analysts though probably due to convenience rather than lack of knowledge. I’m fairly certain most individuals don’t appreciate the differences (since many companies themselves don’t understand). And if the sell-side don’t clarify (or don’t care), I’m guessing most Japanese financial institutions ignore it. And international investors focus on large caps in order to compare with global peers, but, as mentioned, the reconciliations aren’t that clean in Japan where its implementation of IFRS feels more principles-based than the rules-based methodology of J-GAAP, at least compared to the US (though probably less so than Europe which often feels arbitrary to me, not understanding the intricacies of European accounting). Even I didn’t really notice the differences as IFRS implementation is still rather new in Japan and I only noticed after our coverage of large caps increased as well as that of relistings (where private equity generally force IFRS implementation, guessing for the reasons mentioned above).
 
This is yet another reason why there are plenty of opportunities in Japan for stock price dislocations from a valuation perspective alone, especially as the market appears to be even more short-term focused (much of the fine print needed for reconciliations are only available in the annual which is published after the AGM or 1~2 months after full year results are initially released). And being more principles-based, it becomes important to understand how management “thinks” about their financials (if they do at all … I’ve noticed, particularly among relistings, that management themselves don’t appreciate the accounting disparities since IFRS was forced upon them). Just as I can’t easily analyze European financials, I suspect the reverse is also true. When I screen against global comps, I certainly don’t adjust for accounting variations across countries, which probably makes my Japanese companies, who often use J-GAAP, look least profitable vs their global peers. The funny thing about all of this is that “true” value of the business shouldn’t depend on accounting practices. But, in the stock markets, they do and it is very important to be aware of them. Anyway, I hope that at least long-term investors appreciate these distinctions.
 
My current favorite metric that the market seems to focus on is the new EBITDA, “Earnings Before Iran, Tariffs, and Donald Announcements”.
​